Federated Analog Data Pipelines Across Partner Space Agencies
Problem
A modern planetary analog campaign rarely belongs to one space agency anymore, and the data-pipeline burden has not caught up with that reality. The NASA LunaNet Interoperability Specification (2025 baseline) lays out the federated service specification NASA, ESA, JAXA, and partner agencies have agreed to baseline against, and the NASA LunaNet Interoperability Specification program page documents the multi-agency provider model that mission architects now have to design into flight concepts. Federated analog pipelines across partner space agencies have to survive that spec from analog through flight without breaking the provenance chain.
The difficulty is not standards — NASA, ESA, and JAXA have been publishing standards for decades. The difficulty is federating the actual data. The NASA Planetary Data Ecosystem is the umbrella archive on the US side, the ESA Planetary Science Archive is its ESA counterpart, and JAXA runs its own equivalent. An analog pipeline that produces a single dataset should be able to hand that dataset to any of the three archives with minimal rework. In practice, most analog programs end up producing three separate deliverables and doing the federation work three times — which is both expensive and the main reason cross-agency analog papers take years to publish.
A cross-niche parallel from cave-diving illustrates the federation discipline at smaller scale. Our federated expedition data work uses the same federation layer and has turned out to be a useful stress test, since cave-diving teams federate data across multiple national governing bodies with their own archival standards. The cave-diving federation problem is structurally similar to the multi-agency planetary federation problem: multiple equally-authoritative data consumers, each with its own conventions, all wanting access to a single dataset without making the data producer maintain three parallel exports. Methods that work for cave-diving federation transfer directly into planetary federation, and the cross-domain validation has accelerated maturity of both implementations.
Solution
EchoQuilt's federation layer was built to survive LunaNet conformance, CCSDS bundling, and the three-archive reality at once. Each quilt patch is stored with a provenance record that is simultaneously PDS4-conformant, ESA PSA-conformant, and JAXA-ready. The stitching engine packs completed quilts into CCSDS-style bundles whose metadata headers tell each archive what it needs to know — without requiring the analog team to maintain three parallel exporters. The CCSDS Consultative Committee for Space Data Systems standards provide the reference for the bundle format, and EchoQuilt's exporters implement the subset needed for analog deliverables today.
We ran the federation layer against a NASA-ESA-JAXA cross-validation during the 2025 Mauna Loa season. The Hawaiian analog quilt was pushed through the federated pipeline as a single artifact, and each partner archive successfully ingested it without schema rework. The Planetary Society piece on how NASA and ESA work together documents the joint-mission track record that makes this kind of federation viable in the first place — the agencies already know how to collaborate; the pipeline just has to get out of the way.
The cross-validation produced specific operational lessons that fed back into the federation layer's design. PDS4 ingestion required the patch metadata to include a specific spatial coordinate system identifier that ESA PSA and JAXA do not require by default; rather than maintain three exporters, we extended the metadata schema to carry the identifier as an optional field that defaults to a sensible value and gets used by PDS4 when present. ESA PSA's ingestion verified the temporal resolution of patch timestamps against an internal precision standard; we adjusted the patch timestamp format to satisfy ESA's requirement without breaking compatibility with the other archives. JAXA's ingestion accepted the bundle but flagged it for human review on first ingestion; subsequent ingestions cleared automatically once JAXA's automation team had whitelisted the bundle source.
Each of these adjustments was small in isolation but collectively demonstrated that federation in practice requires negotiation, accommodation, and iteration that a single-archive pipeline never has to face.
Three design choices made the federation layer work in practice. First, the quilt schema carries LunaNet-style service identifiers on every patch, which lets any partner relay or archive recognize the data without a separate handshake. Second, the federation exporter is idempotent — pushing the same quilt to the same archive twice produces the same archive record — which matters enormously when a multi-agency review cycle forces re-submissions. Third, the pipeline supports cross-niche borrows: the cross-discipline handoffs work from our own analog teams uses the same exporter path, which validated the federation layer's flexibility for non-mission datasets before we committed to multi-agency analog deployments.
Advanced tactics
Three tactics sharpen the federation layer past the default single-quilt export. First, treat archive ingestion as a deliverable, not an afterthought. Most analog programs schedule data delivery at the end of the campaign, which means archive-specific rework shows up when the crew has already demobilized. EchoQuilt's pipeline produces archive-ready bundles during the campaign, so the first partner archive ingestion usually happens before the last crew has left the field. This single operational change has cut our time-to-archive by roughly 4 months in our last three campaigns.
Second, negotiate the federation schema with at least one partner archive operator before the first campaign field day. The NASA LunaNet Interoperability Specification baselines a lot, but every archive has local conventions, and pre-negotiating them against the LunaNet spec turns a potential six-month schema argument into a one-week alignment. Our own schema reviews with PDS and PSA operators now happen during the proposal phase, not after the first dataset is delivered.
Third, maintain a federation test harness that replays historical quilts against the live exporter on every exporter change. This is straightforward engineering discipline but it catches the kind of quiet regression that otherwise shows up months later in an archive audit. The harness also lets you prove LunaNet conformance continuously, which matters as the spec evolves across its 2025 through 2027 baselines.
Fourth, document successful federation cases as patterns for future campaigns. The Marius analog tests campaign produced the first cross-agency-archived quilt we shipped, and the lessons from that federation have been distilled into a reusable pattern that subsequent multi-agency campaigns can follow. Pattern documentation matters because federation is not a one-time problem — every new campaign brings its own variations on the same themes, and a documented pattern shortcuts the negotiation work that would otherwise have to be redone for each campaign.
Fifth, plan for federation evolution. The LunaNet spec is on a continuous baseline cycle through 2027 and beyond, and partner archives are simultaneously evolving their own ingestion APIs. A federation layer that works well today against the current spec may need to evolve over the next 18-24 months as new spec versions land and as partner archives add new requirements. EchoQuilt's federation layer is designed for incremental evolution rather than wholesale rewrites, so each spec update can be absorbed as a configuration change rather than a code refactor. This evolutionary architecture is the difference between a federation layer that ages well and one that ages quickly.
Sixth, instrument the federation layer with archive-specific telemetry. Each partner archive has its own ingestion-success patterns, and an archive that begins producing more rejected ingests than usual is a signal that something has changed — either in the partner archive's policy or in the federation layer's output. EchoQuilt's federation layer logs per-archive ingestion outcomes and surfaces anomaly alerts when rejection rates drift outside expected bounds. This early-warning capability has caught two partner-archive policy changes ahead of any formal communication from the partner agencies, which let our federation team adapt before user-facing impact.
CTA
If your team is planning a NASA-ESA-JAXA joint analog campaign, a LunaNet-conformant flight concept, or a NIAC proposal that has to deliver to multiple partner archives, EchoQuilt's federation layer is already shipping archive-ready bundles. Each pilot ships with the federation schema baselined against the LunaNet Interoperability Specification 2025 release, the CCSDS bundle exporter that conforms to the Consultative Committee for Space Data Systems standards, the PDS4 and ESA PSA conformance checklist we use on every campaign, the federation test harness that lets you validate your own quilts against partner-archive rules before submission, an idempotent exporter that produces the same archive record across re-submissions during multi-agency review cycles, and per-archive ingestion telemetry that surfaces anomaly alerts when rejection rates drift outside expected bounds.
Pilot teams shape the cross-agency LunaNet service identifier defaults and the evolution-aware federation architecture that the 2027 reference release will adopt as LunaNet baselines roll forward through 2027 and beyond. Priority goes to NIAC PIs preparing multi-agency proposals in the 2026 cycle, JPL mission architects scoping joint NASA-ESA-JAXA campaigns, ESA Planetary Science Archive coordinators integrating analog quilt ingestion, JAXA archive operators preparing automated whitelist workflows, and Artemis-era flight concept teams targeting LunaNet conformance gates. Join the Waitlist for Planetary Analog Researchers and we will share the federation schema, the CCSDS bundle exporter, and the PDS4/PSA conformance checklist we use on every campaign.